## Modelling the future of glaciers in Tirol

My former colleague Ben Marzeion (now at Bremen University) developed a simple way to calculate the changes in glaciers using only air temperature and precipitation as inputs. What is critical to a glacier is how much snowfall it gets per year (which adds mass to the glacier) and how much of the glacier will melt each year. The temperature dictates whether the glacier will get rain or snow, and also the air temperature describes a whole bunch of weather conditions that for many glaciers mean that higher temperatures are associated with more melting.

This simple model is fitted to the available records of glacier changes, and when fitted it recreates the historical glacier changes pretty convincingly.  Ben worked with Marlis Hofer to implement a statistical way of estimating the reliability (error) of the model. How this works is that, if , for example you have 20 years of glacier observations, you can fit the model to all 20 years to get the ‘best’ model result, but you can also fit it to 19 of the 20 years and then assess you well the model reproduces the year that you left out. If you do this assessment by leaving out each of the 20 years in turn you can use the results as an indication of how well your model works for predicting unknown years.

Ok, so by running this model into the future we can get an idea of how the glaciers of Tirol will be up to the end of the century. But of course to do this we first need a set of projected climate conditions. These are taken from the IPCC simulations of future climate, but as we don’t know how humanity is going to behave (carry on emitting greenhouse gases or push for a switch from fossil fuels) various future climates are simulated using ‘representative concentration pathways‘ (RCPs) that correspond to different actions on the part of the people of the world.

So lets look at the example of the Stubaier Alps, closest to Innsbruck (description below figure):

Modelled evolution of the glaciers in the Stubaier Alps (1900 – 2100 AD)  based on different climate models and climate scenarios. On the left axes the charts show (top) cumulative annual change in glacier volume; (middle) annual rate of change of volume and (bottom) cumulative change in area. On the right axes the charts show the glacier volume as a % of the volume in 2000, (middle) annual volume change as a % of the glacier volume as a % of the volume  of the year before (bottom) the glacier area through time as a % of the glacier area in 2000. The black line (uncertainty ranges in grey) is modeled using climate model input from CMIP5 experiment (historical) and in purple using the Climate Research Unit climate reconstruction from available measurements (CRU), to the right of the graph, where things are more colorful, the model is forced by the four possible climate futures used in the IPCC, all of which are considered possible depending on how much greenhouse gases are emitted in the years to come. The four RCPs, RCP2.6, RCP4.5, RCP6, and RCP8.5, are named after a possible range of radiative forcing values in the year 2100 relative to pre-industrial values (essentially how much more energy the climate system will contain relative to now: +2.6, +4.5, +6.0, and +8.5 W/m2, respectively). The numbers at the top (57.8km2 and 2.2km2) are the average glacier area and volume of this area between 1986 and 2005 as obtained by the model.

So, you can see from just the top panel that by 2050 this model predicts that glacier areas are likely to be <25% of their area in 2000, regardless of what levels of greenhouse gases we emit, so not looking good for the Stubaier Alps glaciers is it? But what about the rest of Tirol? Well the figure below shows is that even with the most aggressive reduction in green house gas emissions the glaciers in Tirol will reduce to about 10% of their 2000 size by the end of the century. If we  make do not alter our current behaviour regarding greenhouse gas emissions, its possible that there will be no glaciers left in Tirol by the end of the century. The results of this modelling exercise suggest that the Tirolean glaciers will be practically gone within the lifetime of our children.

The data for these figures comes from: Marzeion, B.; Hofer, M.; Jarosch, A. H.; Kaser, G.; Mölg, T. A Minimal Model for Reconstructing Interannual Mass Balance Variability of Glaciers in the European Alps. The Cryosphere. 2012, 6 (1), 71–84 DOI: 10.5194/tc-6-71-2012.

The figures and description were provided by Wolfgang Gurgiser